fix bug about tensor str in nonsymmetric cast and getitem in consist…

[liufengwei0103]

1. 支持非对称情况下的打印
2. 解决GetItem在判断is_identity进行copy时，consistent tensor缺乏条件的问题。

[github-actions]

Speed stats:

GPU Name: GeForce GTX 1080 

OneFlow resnet50 time: 128.2ms (= 6410.0ms / 50, input_shape=[16, 3, 224, 224])
PyTorch resnet50 time: 140.3ms (= 7012.7ms / 50, input_shape=[16, 3, 224, 224])
✔️ Relative speed: 1.09 (= 140.3ms / 128.2ms)

OneFlow resnet50 time: 76.6ms (= 3828.0ms / 50, input_shape=[8, 3, 224, 224])
PyTorch resnet50 time: 85.1ms (= 4253.1ms / 50, input_shape=[8, 3, 224, 224])
✔️ Relative speed: 1.11 (= 85.1ms / 76.6ms)

OneFlow resnet50 time: 51.1ms (= 2554.4ms / 50, input_shape=[4, 3, 224, 224])
PyTorch resnet50 time: 61.1ms (= 3052.6ms / 50, input_shape=[4, 3, 224, 224])
✔️ Relative speed: 1.20 (= 61.1ms / 51.1ms)

OneFlow resnet50 time: 44.8ms (= 2237.8ms / 50, input_shape=[2, 3, 224, 224])
PyTorch resnet50 time: 50.9ms (= 2545.7ms / 50, input_shape=[2, 3, 224, 224])
✔️ Relative speed: 1.14 (= 50.9ms / 44.8ms)

OneFlow resnet50 time: 45.0ms (= 2249.7ms / 50, input_shape=[1, 3, 224, 224])
PyTorch resnet50 time: 44.0ms (= 2199.1ms / 50, input_shape=[1, 3, 224, 224])
✔️ Relative speed: 0.98 (= 44.0ms / 45.0ms)

OneFlow resnet50 time: 156.8ms (= 7840.5ms / 50, input_shape=[16, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 159.8ms (= 7990.7ms / 50, input_shape=[16, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.02 (= 159.8ms / 156.8ms)

OneFlow resnet50 time: 98.7ms (= 4933.1ms / 50, input_shape=[8, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 102.6ms (= 5128.4ms / 50, input_shape=[8, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.04 (= 102.6ms / 98.7ms)

OneFlow resnet50 time: 81.9ms (= 4096.4ms / 50, input_shape=[4, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 74.5ms (= 3725.2ms / 50, input_shape=[4, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 0.91 (= 74.5ms / 81.9ms)

OneFlow resnet50 time: 71.9ms (= 3596.4ms / 50, input_shape=[2, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 74.0ms (= 3699.3ms / 50, input_shape=[2, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.03 (= 74.0ms / 71.9ms)

OneFlow resnet50 time: 67.5ms (= 3374.2ms / 50, input_shape=[1, 3, 224, 224], ddp, world size=2)
PyTorch resnet50 time: 69.1ms (= 3457.2ms / 50, input_shape=[1, 3, 224, 224], ddp, world size=2)
✔️ Relative speed: 1.02 (= 69.1ms / 67.5ms)